Klasifikasi Sinyal Eeg Menggunakan Discrete Wavelet Transform (DWT) Dan Gaussian Support Vector Machine (SVM) Untuk Penderita Epilepsi

Ijma’u, Rizki (2021) Klasifikasi Sinyal Eeg Menggunakan Discrete Wavelet Transform (DWT) Dan Gaussian Support Vector Machine (SVM) Untuk Penderita Epilepsi. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text 1. Cover.pdf Download (667kB)
	Text 3. Abstract.pdf Download (205kB)
	Text 2. Abstrak.pdf Download (119kB)
	Text 4. Bab 1.pdf Download (225kB)
	Text 5. Bab 2.pdf Download (541kB)
	Text 6. Bab 3.pdf Download (327kB)
	Text 7. Bab 4.pdf Restricted to Registered users only Download (164kB) | Request a copy
	Text 8. Bab 5.pdf Download (6kB)
	Text 9. Daftar Pustaka.pdf Download (130kB)
	Text 10. Lampiran.pdf Restricted to Registered users only Download (1MB) | Request a copy

Abstract

Epileptic is classified as a neurological disorder that is considered dangerous with a high probability of death. Epileptic sufferers are advised from an early age to get proper treatment so that their condition does not worsen. One of the proper handling is performing detection based on Electroenchephalogram (EEG) technology. The EEG helps differentiate between normal and epileptic patients. EEG recording data used for this study were obtained online from “Epileptology Clinic, University of Bonn”. Data presented in the form of 5 conditions originating from volunteers (normal) and patients with epilepsy with different activities. The data obtained is designed and tested using a system in order to get a good accuracy value. The design of the EEG system is made through several stages including input dataset, decomposition, minimum feature extraction, maximum, mean, standard deviation, kurtosis and skewness of the wavelet coefficients generated from the DWT process and Support Vector Machine (SVM) classification. The method used is in the form of Discrete Wavelet Transform (DWT) with discrete wavelet types in the form of daubechies and SVM classification with Gaussian types. System testing is obtained from the K-Fold process and the Confusion Matrix. The best accuracy results obtained from the condition of volunteers with open and closed eye activity and epilepsy patients with brain activity during interictal and ictal (C1: AB-CDE) DWT and SVM methods can be concluded that there is a feature extraction of a maximum type of 78.6% and a type of db2 of 78.04%. Keywords: epileptic, eeg, dwt, svm

Actions (login required)

View Item